treatment dataset_anon samples
Hugo2016 PD1 Dataset 1 28
Prat2017 PD1 Dataset 2 25
Riaz2017 PD1 Dataset 3 56
TCGA CTLA4 Dataset 4 19
VanAllen2015 CTLA4 Dataset 5 42

Scale data by dataset mean and standard deviation

Test gene list for differences in response

Just plot a couple of genes of interest for sitc presentation

Of 8 genes of interest, 8 are represented by gene symbol in the IMPRES set.

For each of those, compute ANOVA for linear model, with dataset and objective_response as effects.

P-values for the genes of interest
P-value
PDCD1 0.06440
GZMB 0.06620
LAG3 0.06985
PRF1 0.07895
CD274 0.09495
CD8A 0.09949
CTLA4 0.13931
HAVCR2 0.20372

Plots for each gene

Longer list

Of 715 genes of interest, 551 are represented by gene symbol in the IMPRES set.

For each of those, compute ANOVA for linear model, with dataset and objective_response as effects.

Lowest p-values

P-value
CYFIP2 0.00389
CCR7 0.01251
NFKB2 0.01308
ST6GAL1 0.01470
CXCL13 0.01984
TNFRSF11B 0.02134
STAT3 0.02381
HSD11B1 0.02519
IL21R 0.03024
IL2RA 0.03321
CD3E 0.03585
SELE 0.03623
BCL6 0.03764
CARD11 0.04001
FLT3 0.04090
IL17RA 0.04130
IRF1 0.04222
NFKB1 0.04280
TLR8 0.04307
LTB 0.04324

Plots for top genes

More focused list

Of 34 genes of interest, 27 are represented by gene symbol in the IMPRES set.

For each of those, compute ANOVA for linear model, with dataset and objective_response as effects.

P-values for the genes of interest
P-value
TIGIT 0.04394
CXCL9 0.05323
CXCR3 0.06164
PDCD1 0.06440
GZMB 0.06620
LAG3 0.06985
IDO1 0.07000
PRF1 0.07895
CD274 0.09495
CD8A 0.09949
IFNGR1 0.11839
TNFRSF9 0.12022
IRF8 0.13908
CXCL10 0.14721
THBD 0.16916
GZMA 0.18311
HAVCR2 0.20372
CCL5 0.23928
FLT3LG 0.32396
IFNGR2 0.34578
ITGAM 0.50607
XCL1 0.52420
BATF3 0.55156
NT5E 0.63707
ENTPD1 0.74859
TGFB1 0.78055
ITGAE 0.82085

Plots for each gene